Tester
Lipp Laboratory, Department of Environmental Health Science, University of Georgia
2020-10-21
Load all required libraries.
library(tidyverse)## Warning: package 'tidyverse' was built under R version 3.6.3## -- Attaching packages ------------------------------------------------------------------- tidyverse 1.3.0 --## v ggplot2 3.3.2 v purrr 0.3.4
## v tibble 3.0.3 v dplyr 1.0.0
## v tidyr 1.1.0 v stringr 1.4.0
## v readr 1.3.1 v forcats 0.5.0## Warning: package 'ggplot2' was built under R version 3.6.3## Warning: package 'tibble' was built under R version 3.6.3## Warning: package 'readr' was built under R version 3.6.3## Warning: package 'dplyr' was built under R version 3.6.3## Warning: package 'forcats' was built under R version 3.6.3## -- Conflicts ---------------------------------------------------------------------- tidyverse_conflicts() --
## x dplyr::filter() masks stats::filter()
## x dplyr::lag() masks stats::lag()library(plotly)## Warning: package 'plotly' was built under R version 3.6.3##
## Attaching package: 'plotly'## The following object is masked from 'package:ggplot2':
##
## last_plot## The following object is masked from 'package:stats':
##
## filter## The following object is masked from 'package:graphics':
##
## layoutlibrary(broom)## Warning: package 'broom' was built under R version 3.6.3Read in raw data from RDS.
raw_data <- readRDS("./n1_n2_cleaned_cases.rds")Make a few small modifications to names and data for visualizations.
final_data <- raw_data %>% mutate(log_copy_per_L = log10(mean_copy_num_L)) %>%
rename(Facility = wrf) %>%
mutate(Facility = recode(Facility,
"NO" = "WRF A",
"MI" = "WRF B",
"CC" = "WRF C"))Seperate the data by gene target to ease layering in the final plot
#make three data layers
only_positives <<- subset(final_data, (!is.na(final_data$Facility)))
only_n1 <- subset(only_positives, target == "N1")
only_n2 <- subset(only_positives, target == "N2")
only_background <<-final_data %>%
select(c(date, cases_cum_clarke, new_cases_clarke, X7_day_ave_clarke, cases_per_100000_clarke)) %>%
group_by(date) %>% summarise_if(is.numeric, mean)
#specify fun colors
background_color <- "#7570B3"
seven_day_ave_color <- "#E6AB02"
marker_colors <- c("N1" = '#1B9E77',"N2" ='#D95F02')#remove facilty C for now
only_n1 <- only_n1[!(only_n1$Facility == "WRF C"),]
only_n2 <- only_n2[!(only_n2$Facility == "WRF C"),]Build the main plot
#first layer is the background epidemic curve
p1 <- only_background %>%
plotly::plot_ly() %>%
plotly::add_trace(x = ~date, y = ~new_cases_clarke,
type = "bar",
hoverinfo = "text",
text = ~paste('</br> Date: ', date,
'</br> Daily Cases: ', new_cases_clarke),
alpha = 0.5,
name = "Daily Reported Cases",
color = background_color,
colors = background_color,
showlegend = FALSE) %>%
layout(yaxis = list(title = "Clarke County Daily Cases", showline=TRUE)) %>%
layout(legend = list(orientation = "h", x = 0.2, y = -0.3))
#renders the main plot layer two as seven day moving average
p1 <- p1 %>% plotly::add_trace(x = ~date, y = ~X7_day_ave_clarke,
type = "scatter",
mode = "lines",
hoverinfo = "text",
text = ~paste('</br> Date: ', date,
'</br> Seven-Day Moving Average: ', X7_day_ave_clarke),
name = "Seven Day Moving Average Athens",
line = list(color = seven_day_ave_color),
showlegend = FALSE)
#renders the main plot layer three as positive target hits
p2 <- plotly::plot_ly() %>%
plotly::add_trace(x = ~date, y = ~mean_copy_num_L,
type = "scatter",
mode = "markers",
hoverinfo = "text",
text = ~paste('</br> Date: ', date,
'</br> Facility: ', Facility,
'</br> Target: ', target,
'</br> Copies/L: ', round(mean_copy_num_L, digits = 2)),
data = only_n1,
symbol = ~Facility,
marker = list(color = '#1B9E77', size = 8, opacity = 0.65),
showlegend = FALSE) %>%
plotly::add_trace(x = ~date, y = ~mean_copy_num_L,
type = "scatter",
mode = "markers",
hoverinfo = "text",
text = ~paste('</br> Date: ', date,
'</br> Facility: ', Facility,
'</br> Target: ', target,
'</br> Copies/L: ', round(mean_copy_num_L, digits = 2)),
data = only_n2,
symbol = ~Facility,
marker = list(color = '#D95F02', size = 8, opacity = 0.65),
showlegend = FALSE) %>%
layout(yaxis = list(title = "SARS CoV-2 Copies/L",
showline = TRUE,
type = "log",
dtick = 1,
automargin = TRUE)) %>%
layout(legend = list(orientation = "h", x = 0.2, y = -0.3))
#adds the limit of detection dashed line
p2 <- p2 %>% plotly::add_segments(x = as.Date("2020-03-14"),
xend = ~max(date + 10),
y = 3571.429, yend = 3571.429,
opacity = 0.35,
line = list(color = "black", dash = "dash")) %>%
layout(annotations = list(x = as.Date("2020-03-28"), y = 3.8, xref = "x", yref = "y",
text = "Limit of Detection", showarrow = FALSE))
p1## Warning: `arrange_()` is deprecated as of dplyr 0.7.0.
## Please use `arrange()` instead.
## See vignette('programming') for more help
## This warning is displayed once every 8 hours.
## Call `lifecycle::last_warnings()` to see where this warning was generated.## Warning: Ignoring 1 observations p2## Warning: `group_by_()` is deprecated as of dplyr 0.7.0.
## Please use `group_by()` instead.
## See vignette('programming') for more help
## This warning is displayed once every 8 hours.
## Call `lifecycle::last_warnings()` to see where this warning was generated.p_combined <-
plotly::subplot(p2,p1, # plots to combine, top to bottom
nrows = 2,
heights = c(.6,.4), # relative heights of the two plots
shareX = TRUE, # plots will share an X axis
titleY = TRUE
) %>%
# create a vertical "spike line" to compare data across 2 plots
plotly::layout(
xaxis = list(
spikethickness = 1,
spikedash = "dot",
spikecolor = "black",
spikemode = "across+marker",
spikesnap = "cursor"
),
yaxis = list(spikethickness = 0)
)## Warning: Ignoring 1 observationsp_combinedSave the plot to pull into the index
save(p_combined, file = "./plotly_fig.rda")Save an htmlwidget for website embedding
htmlwidgets::saveWidget(p_combined, "plotly_fig.html")Build loess smoothing figures figures
#create smoothing data frames
#n1
smooth_n1 <- only_n1 %>% select(-c(Facility)) %>%
group_by(date, cases_cum_clarke, new_cases_clarke, X7_day_ave_clarke, cases_per_100000_clarke) %>%
summarize(sum_copy_num_L = sum(mean_total_copies)) %>%
ungroup() %>%
mutate(log_sum_copies_L = log10(sum_copy_num_L)) %>%
mutate(target = "N1")## `summarise()` regrouping output by 'date', 'cases_cum_clarke', 'new_cases_clarke', 'X7_day_ave_clarke' (override with `.groups` argument)#n2
smooth_n2 <- only_n2 %>% select(-c(Facility)) %>%
group_by(date, cases_cum_clarke, new_cases_clarke, X7_day_ave_clarke, cases_per_100000_clarke) %>%
summarize(sum_copy_num_L = sum(mean_total_copies)) %>%
ungroup() %>%
mutate(log_sum_copies_L = log10(sum_copy_num_L)) %>%
mutate(target = "N2")## `summarise()` regrouping output by 'date', 'cases_cum_clarke', 'new_cases_clarke', 'X7_day_ave_clarke' (override with `.groups` argument)#add trendlines
#extract data from geom_smooth
#n1 extract
# *********************************span 0.6***********************************
#*****************Must always update the n = TOTAL NUMBER OF DAYS*************************
extract_n1 <- ggplot(smooth_n1, aes(x = date, y = log_sum_copies_L)) +
stat_smooth(aes(outfit=fit_n1<<-..y..), method = "loess", color = '#1B9E77',
span = 0.6, n = 127)## Warning: Ignoring unknown aesthetics: outfit#n2 extract
extract_n2 <- ggplot(smooth_n2, aes(x = date, y = log_sum_copies_L)) +
stat_smooth(aes(outfit=fit_n2<<-..y..), method = "loess", color = '#1B9E77',
span = 0.6, n = 127)## Warning: Ignoring unknown aesthetics: outfit#look at the fits to align dates and total observations
#n1
extract_n1## `geom_smooth()` using formula 'y ~ x'
fit_n1## [1] 10.89030 11.08191 11.26703 11.44538 11.61665 11.78053 11.93672 12.08492
## [9] 12.22494 12.35706 12.48171 12.59931 12.71027 12.81503 12.91399 13.00584
## [17] 13.08920 13.16440 13.23180 13.29174 13.34454 13.39057 13.43015 13.46363
## [25] 13.49135 13.51366 13.53088 13.54337 13.55147 13.54550 13.51768 13.47120
## [33] 13.40926 13.33503 13.25171 13.16249 13.07055 12.97909 12.89130 12.81036
## [41] 12.73946 12.68179 12.64055 12.60007 12.54689 12.48742 12.42809 12.37535
## [49] 12.33560 12.31530 12.31114 12.31515 12.32672 12.34525 12.37012 12.40073
## [57] 12.43649 12.47677 12.52097 12.56849 12.61872 12.67106 12.72489 12.77962
## [65] 12.84310 12.92177 13.01290 13.11376 13.22161 13.33373 13.44737 13.55981
## [73] 13.66831 13.77015 13.86258 13.94287 14.00829 14.05612 14.09422 14.13136
## [81] 14.16614 14.19716 14.22301 14.24230 14.25362 14.25284 14.23804 14.21108
## [89] 14.17386 14.12823 14.07609 14.01931 13.95977 13.89933 13.83989 13.78332
## [97] 13.73149 13.68629 13.64959 13.61610 13.57970 13.54099 13.50058 13.45909
## [105] 13.41711 13.37527 13.33417 13.29441 13.25662 13.22139 13.18933 13.16106
## [113] 13.13719 13.11634 13.09679 13.07856 13.06173 13.04632 13.03239 13.01999
## [121] 13.00916 12.99995 12.99241 12.98659 12.98253 12.98028 12.97989#n2
extract_n2## `geom_smooth()` using formula 'y ~ x'
fit_n2## [1] 10.71559 10.93508 11.14769 11.35320 11.55139 11.74205 11.92494 12.09986
## [9] 12.26663 12.42543 12.57658 12.72038 12.85715 12.98721 13.11085 13.22729
## [17] 13.33565 13.43614 13.52895 13.61428 13.69233 13.76330 13.82739 13.88480
## [25] 13.93573 13.98038 14.01895 14.05163 14.07864 14.09170 14.08425 14.05899
## [33] 14.01863 13.96590 13.90351 13.83416 13.76058 13.68549 13.61158 13.54159
## [41] 13.47821 13.42418 13.38219 13.33651 13.27439 13.20306 13.12974 13.06165
## [49] 13.00602 12.97008 12.94838 12.93067 12.91695 12.90723 12.90150 12.89978
## [57] 12.90207 12.90836 12.91867 12.93300 12.95136 12.97374 13.00015 13.03059
## [65] 13.07140 13.12719 13.19536 13.27334 13.35853 13.44836 13.54024 13.63158
## [73] 13.71981 13.80232 13.87654 13.93989 13.98978 14.02362 14.05163 14.08360
## [81] 14.11630 14.14650 14.17098 14.18649 14.18982 14.17771 14.15052 14.11041
## [89] 14.05953 14.00003 13.93408 13.86383 13.79144 13.71905 13.64884 13.58296
## [97] 13.52355 13.47279 13.43282 13.39746 13.35954 13.31974 13.27874 13.23724
## [105] 13.19590 13.15541 13.11646 13.07973 13.04590 13.01565 12.98966 12.96863
## [113] 12.95322 12.94194 12.93289 12.92610 12.92164 12.91957 12.91992 12.92277
## [121] 12.92816 12.93615 12.94680 12.96015 12.97627 12.99520 13.01700#assign fits to a vector
n1_trend <- fit_n1
n2_trend <- fit_n2
#extract y min and max for each
limits_n1 <- ggplot_build(extract_n1)$data## `geom_smooth()` using formula 'y ~ x'limits_n1 <- as.data.frame(limits_n1)
n1_ymin <- limits_n1$ymin
n1_ymax <- limits_n1$ymax
limits_n2 <- ggplot_build(extract_n2)$data## `geom_smooth()` using formula 'y ~ x'limits_n2 <- as.data.frame(limits_n2)
n2_ymin <- limits_n2$ymin
n2_ymax <- limits_n2$ymax
#reassign dataframes (just to be safe)
work_n1 <- smooth_n1
work_n2 <- smooth_n2
#fill in missing dates to smooth fits
work_n1 <- work_n1 %>% complete(date = seq(min(date), max(date), by = "1 day"))
date_vec_n1 <- work_n1$date
work_n2 <- work_n2 %>% complete(date = seq(min(date), max(date), by = "1 day"))
date_vec_n2 <- work_n2$date
#create a new smooth dataframe to layer
smooth_frame_n1 <- data.frame(date_vec_n1, n1_trend, n1_ymin, n1_ymax)
smooth_frame_n2 <- data.frame(date_vec_n2, n2_trend, n2_ymin, n2_ymax)#make plotlys
#plot smooth frames
p3 <- plotly::plot_ly() %>%
plotly::add_lines(x = ~date_vec_n1, y = ~n1_trend,
data = smooth_frame_n1,
hoverinfo = "text",
text = ~paste('</br> Date: ', date_vec_n1,
'</br> Median Log Copies: ', round(n1_trend, digits = 2),
'</br> Target: N1'),
line = list(color = '#1B9E77', size = 8, opacity = 0.65),
showlegend = FALSE) %>%
plotly::add_lines(x = ~date_vec_n2, y = ~n2_trend,
data = smooth_frame_n2,
hoverinfo = "text",
text = ~paste('</br> Date: ', date_vec_n2,
'</br> Median Log Copies: ', round(n2_trend, digits = 2),
'</br> Target: N2'),
line = list(color = '#D95F02', size = 8, opacity = 0.65),
showlegend = FALSE) %>%
plotly::add_ribbons(x ~date_vec_n1, ymin = ~n1_ymin, ymax = ~n1_ymax,
showlegend = FALSE,
opacity = 0.25,
hoverinfo = "text",
text = ~paste('</br> Date: ', date_vec_n1, #leaving in case we want to change
'</br> Max Log Copies: ', round(n1_ymax, digits = 2),
'</br> Min Log Copies: ', round(n1_ymin, digits = 2),
'</br> Target: N1'),
name = "",
line = list(color = '#1B9E77')) %>%
plotly::add_ribbons(x ~date_vec_n2, ymin = ~n2_ymin, ymax = ~n2_ymax,
showlegend = FALSE,
opacity = 0.25,
hoverinfo = "text",
text = ~paste('</br> Date: ', date_vec_n2, #leaving in case we want to change
'</br> Max Log Copies: ', round(n2_ymax, digits = 2),
'</br> Min Log Copies: ', round(n2_ymin, digits = 2),
'</br> Target: N2'),
name = "",
line = list(color = '#D95F02')) %>%
layout(yaxis = list(title = "Total Log SARS CoV-2 Copies",
showline = TRUE,
automargin = TRUE)) %>%
layout(xaxis = list(title = "Date")) %>%
plotly::add_segments(x = as.Date("2020-06-24"),
xend = as.Date("2020-06-24"),
y = ~min(n1_ymin), yend = ~max(n1_ymax),
opacity = 0.35,
name = "Bars Repoen",
hoverinfo = "text",
text = "</br> Bars Reopen",
"</br> 2020-06-24",
showlegend = FALSE,
line = list(color = "black", dash = "dash")) %>%
plotly::add_segments(x = as.Date("2020-07-09"),
xend = as.Date("2020-07-09"),
y = ~min(n1_ymin), yend = ~max(n1_ymax),
opacity = 0.35,
name = "Mask Mandate",
hoverinfo = "text",
text = "</br> Mask Mandate",
"</br> 2020-07-09",
showlegend = FALSE,
line = list(color = "black", dash = "dash")) %>%
plotly::add_segments(x = as.Date("2020-08-20"),
xend = as.Date("2020-08-20"),
y = ~min(n1_ymin), yend = ~max(n1_ymax),
opacity = 0.35,
name = "</br> Classes Begin",
"</br> 2020-08-20",
hoverinfo = "text",
text = "Classes Begin",
showlegend = FALSE,
line = list(color = "black", dash = "dash")) %>%
plotly::add_segments(x = as.Date("2020-10-03"),
xend = as.Date("2020-10-03"),
y = ~min(n1_ymin), yend = ~max(n1_ymax),
opacity = 0.35,
name = "</br> First Home Football Game",
"</br> 2020-10-03",
hoverinfo = "text",
text = "First Home Football Game",
showlegend = FALSE,
line = list(color = "black", dash = "dash")) %>%
plotly::add_markers(x = ~date, y = ~log_sum_copies_L,
data = smooth_n1,
hoverinfo = "text",
showlegend = FALSE,
text = ~paste('</br> Date: ', date,
'</br> Actual Log Copies: ', round(log_sum_copies_L, digits = 2)),
marker = list(color = '#1B9E77', size = 6, opacity = 0.65)) %>%
plotly::add_markers(x = ~date, y = ~log_sum_copies_L,
data = smooth_n2,
hoverinfo = "text",
showlegend = FALSE,
text = ~paste('</br> Date: ', date,
'</br> Actual Log Copies: ', round(log_sum_copies_L, digits = 2)),
marker = list(color = '#D95F02', size = 6, opacity = 0.65))
p3Create final trend plot by stacking with epidemic curve
smooth_extract <-
plotly::subplot(p3,p1, # plots to combine, top to bottom
nrows = 2,
heights = c(.6,.4), # relative heights of the two plots
shareX = TRUE, # plots will share an X axis
titleY = TRUE
) %>%
# create a vertical "spike line" to compare data across 2 plots
plotly::layout(
xaxis = list(
spikethickness = 1,
spikedash = "dot",
spikecolor = "black",
spikemode = "across+marker",
spikesnap = "cursor"
),
yaxis = list(spikethickness = 0)
)## Warning: Ignoring 1 observationssmooth_extractsave(smooth_extract, file = "./smooth_extract.rda")